The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Bash to execute system commands and Node.js scripts for diagnostic purposes. It runs node --check to validate hook script syntax and suggests running node scripts/install-apply.js for environment repair. It also executes internal diagnostic logic from ~/.claude/scripts/lib/env-doctor-checks.js.
[DATA_EXPOSURE]: The skill accesses local configuration files and session directories to verify environment integrity. This includes reading settings.json for hook registrations and security settings, as well as scanning ~/.claude/ for session files, baseline configurations, and learned skills.
[INDIRECT_PROMPT_INJECTION]: The skill has a potential surface for indirect prompt injection as it processes project files and configurations. Ingestion points include settings.json, agents/*.md, and various skill directories. While it lacks explicit boundary markers or sanitization logic in the description, its capabilities are scoped to environment diagnostics, and the operations are consistent with its primary purpose.

env-doctor